Project Summary Most patients who die from cancer have progression of therapy resistant tumors. Emergent resistance to diverse treatments with distinct mechanisms of activity, termed multidrug resistance, is the greatest barrier to cure yet its causes remain largely unknown. A role for cancer mitochondria in therapy resistance has been sought since these organelles integrate stress and survival signals to determine cell fate. Indeed, most effective cancer therapies induce stress signals sufficient to activate mitochondrial apoptotic signaling, while alterations that repress this process are selected for during tumor progression. To interrogate this directly we optimized an innovative assay in which functional mitochondria are isolated from cancer cells and selectively exposed to tBid and/or Bim, the death stimuli delivered to mitochondria in response to most therapeutic stressors. This provides a read-out of their relative stress sensitivity. We use this tool to study resistance in neuroblastoma, a highly lethal cancer that often completely regresses in response to chemoradiotherapy before subsequently relapsing as multidrug resistant disease. We created a national infrastructure to derive tumor cell lines and patient-derived xenografts from the same patients both at the time of initial diagnosis (before therapy) and again at the time of relapse after treatment. These near-isogenic tumor pairs provide a unique resource as the post-relapse tumors manifest profound multidrug resistance that has been selected for during the course of intensive multimodality treatment. Applying our mitochondrial profiling technique to these tumors enabled the discovery that mitochondria derived from post-relapse therapy resistant tumors have severely blunted apoptotic signaling in response to tBid and Bim in comparison with therapy sensitive tumors. The objective of our work here is to identify the mitochondrial determinants of this therapy resistance. Our central hypothesis based on our preliminary data is that a loss of physical tethering between endoplasmic reticulum and mitochondria is the principal driver of multidrug resistance. ER mitochondria tethers (also termed mitochondria-associated ER membranes, or MAMs) form IP3R/GRP75/VDAC-enriched domains to transfer calcium to mitochondria, and their absence attenuates apoptotic signaling. To test this we will quantify ER- mitochondria contacts in tumors with chemotherapy and kinase inhibitor resistance, manipulate tethering using genetic and biochemical approaches and assess its impact on mitochondrial activities and drug resistance (Aim 1), and define the role calcium plays in this phenotype (Aim 2). While this novel resistance mechanism provides a survival bias downstream of diverse stressors it is not exclusive to other resistance mechanisms. Of note, altered ER-mitochondria tethering has been implicated in diabetes and neurodegeneration as well so its deregulation has broad relevance to human health. The outcomes of these Aims will reveal the contributions of this ER-mitochondria phenotype to cancer therapy resistance, a novel model for the development of tools to measure this, and the identification of therapeutic opportunities to revert resistance.